This is a competing continuation of a RO1 project IRS-1 JCV T-antigen interaction in Cerebellar Tumors. The results from previous founding period demonstrate that Insulin Receptor Substrate 1 (IRS-1), which is the major cytosolic component of the IGF-I receptor signaling system, can translocate to the nucleus. Importantly, we have detected nuclear IRS-1 in medulloblastoma cell lines and in medulloblastoma clinical samples, which were positive for polyomavirus JC large T-antigen. Our further studies demonstrate that nuclear IRS-1 can inhibit faithful component of DNA repair (homologous recombination DNA repair, HRR) via a direct interaction between IRS-1 and Rad51 detected at the sites of damaged DNA. We also demonstrated that the interference with DNA repair by nuclear IRS-1, and subsequent accumulation of mutations may happen in the absence of JCV T-antigen. In this respect, estrogen receptor beta (ER), which is highly expressed in medulloblastomas, can translocate IRS-1 to the nucleus in JCV T-antigen negative cells. These findings indicate that the interaction between IRS-1 and ER could be engaged in cellular responses to DNA damage. It also suggests that inhibition of the faithful DNA repair by nuclear IRS-1 is a common event, not restricted to the medulloblastoma cases which are JCV positive. Therefore, we have developed a research plan which will analyze fundamental mechanisms leading to the development of genomic instability in medulloblastomas. The proposal is founded on the hypothesis that highly active ER found in medulloblastoma mediates translocation of IRS-1 to the nucleus causing inhibition of HRR, unfaithful DNA repair, and the development of genomic instability in actively growing medulloblastoma cells. To test this hypothesis, we have developed a series of experiments, which are outlined in three Specific Aims. In Aim #1, we will evaluate ER effects on IGF-I dependent and IGF-I -independent components of HRR, and will characterize binding and subcellular localizations among ER, IRS-1 and Rad51 in human medulloblastoma cell lines. In Aim#2, we will evaluate how pharmacological and molecular manipulations with the IGF-IR-IRS-1-E signaling axis affect DNA repair fidelity in vitro and tumor progression in RCAS/tv-a transgenic mice. Finally, in Aim #3, we will analyze protein levels, subcellular localization and co-localization among IRS-1, ER, and Rad51 in clinical samples obtained from patients with medulloblastoma. Since DNA damaging agents are frequently used to eliminate CNS tumors including medulloblastoma, defects in DNA repair fidelity may have strong mutagenic consequences. This in turn may result in tumor progression or tumor recurrences after genotoxic treatment. Better understanding of the signaling cross-talk between IRS-1, ER and DNA repair proteins is expected to provide new molecular targets for future therapeutic strategies against genomic instability, which is common in medulloblastoma patients. PUBLIC HEALTH RELEVANCE Medulloblastomas are cerebellar tumors of the childhood, in which the spread of tumor cells within central nervous system (CNS) is associated with poor prognosis. Since DNA damaging agents are frequently used to eliminate medulloblastoma, defects in DNA repair mechanisms are strongly suspected to cause accumulation of mutations contributing to tumor progression or tumor recurrences. The proposed studies of DNA repair mechanisms in association with cell signaling pathways are critically important for the development of new therapeutic strategies against malignant dissemination of these cerebellar tumors in CNS.